Aura H2O Kobo H2O battery life test with light at different settings

[Erich Zann]

During the last weeks, I spent most of my free time in reading. I also tried to evaluate the battery life of the H2O reader I own and check how it changes according to the light intensity. The statistic function of the device is really helpful for that purpose.

I made four tests. Each time I started with a fully loaded battery, Wi-Fi off, set the light to a fixed value and never changed it. I tried to read ~6 hours per day.

• light at 0%: 77,7 hours of reading (during 12 days, average 6,47h/d)
• light at 15%: 62,0 hours of reading (during 13 days, average 4,76h/d)
• light at 50%: 35,9 hours of reading (during 7 days, average 5,12h/d)
• light at 100%: 17,6 hours of reading (during 3 jours, average 5,86h/d)

According to the manufacturer, customers can expect two months of reading based on 30 minutes/day. It equals to 30 hours of reading (and 59 days of standby which seems to eat a signifiant amount of total battery) which is really nice. But if you read much more than 30 minutes, the total reading time is highly increased! Reading for more than 75 hours without charging the device is really impressive in my opinion. And a real asset of ereader against LCD tablets.

Using the frontlight does not necessary decrease by a significant amount the battery life. With an intensity of 15% the battery decrease by 20%. And with 50% intensity the battery life is inferior by 54%. Therefore a linear relation between light intensity and battery life seems to exist. I guess that using the frontlight at low intensity won't significantly impact the total reading time.

But reading with light at 100% (for whitening the screen in daylight) eats a big amount of power, and the total reading time is decreased by 78% compared to an unlit screen!

Last but not least: making such tests is really, really exhausting I hope you'll appreciate it
_____

These values can be compared to what Julius Caesar did some time ago with it Kindle Paperwhite.


Additional test setting:
— latest firmware used (3.19.5761)
— no Wi-Fi
— KePub format
— ~25 lines per page
— page refresh: 128 pages (patch)
— dictionary use: moderate
— device put in sleep instead of shutting it down

[meeera]

Wow, better you than me! I can't stand reading at 75%, let alone 100%.

Did you count page turns at all? Probably a drop in the ocean considering those blaring light levels, and not necessary unless you're reading very different genres/difficulty levels, but just wondering.

[eenk]

Now that's a really useful post! Albeit I personally wouldn't trust the reading statistics unless they have been tested in itself, your overall results seem to be fine. I'm constantly reading with the back light on at around 15% and the battery carries on for quite some weeks.

I don't think that full page refreshes eat too much into the power budget. Your figures about back light eating linearly into the power budget make sense as I would assume a simple pulse-width driven brightness control for the back light LEDs.

The official figures advertised by Kobo need to have a lot of safety margin in order to avoid greedy lawyers trying to extort money from nitpicking.

[Erich Zann]

Thanks for the positive replies

Originally Posted by meeera

Wow, better you than me! I can't stand reading at 75%, let alone 100%.

Did you count page turns at all? Probably a drop in the ocean considering those blaring light levels, and not necessary unless you're reading very different genres/difficulty levels, but just wondering.

I can't read either with frontlight at 100% in a dark room. I had to perform the test with daylight, and during night sessions I really had to enlight the room. Even there it wasn't a pleasant experience (eyeburn).

I haven't count page turns at all. It didn't come to my mind. I must say that I don't really like the idea of measuring battery life as page turns: the amount of signs can vary a lot from one user to another (especially with the amount of tweaking the Kobo allows). Time, on the other side, is universal.

Originally Posted by eenk

Now that's a really useful post! Albeit I personally wouldn't trust the reading statistics unless they have been tested in itself, your overall results seem to be fine.

I checked once if 60 minutes of non-stop reading increased the counter to 1,0 hour. It was the case, so I guess that the statistic counter is working accurately.
But even if it's not perfect, the generated values are still informative and useful to measure the relative impact of light intensity on total battery life.

Another objective criterion is the number of books I was able to fully read between two charges. With 10 easy-reading litterature (400…500 printed pages) finished with no frontlight, I guess it comes close to 80 hours

About page refresh I don't know how much it impacts battery life. I put it to 128 pages because my H2O keeps a good quality display (even if Regal technology is not implemented in this device) to my eyes. There are some artifacts but overall quality is OK to me. I kept this setting for my test.

I also wonder if KePub reading increases the battery life over regular ePub. The formats are nearly the same, but the engine on the Kobo is different. ePub page turns is a bit slower (and supposed to have a bit more typographic accuracy). So I wonder if it requires a bit more CPU power to compute the page (hence the additional time), which should negatively impact the battery life on ~10.000 iterations.

[oren64]

I get almost the same result when i test my Kobo Glo, battery 1200 mAh.

light at 0% (70% of time) light at ~10%(30% of time): ~85 hours of reading
light at 15% (average): ~65 hours of reading

Setting:
— firmware 3.13.1
— no Wi-Fi
— KePub format
— ~28 lines per page
— page refresh: 10 pages (patch)
— dictionary use: ~2 per page
— device power off at night.

Kobo Glo battery specs:
Over 1 month with Wi Fi turned off
Up to 70 hours of continuous usage with ComfortLight on

When the device is shut off when the battrey is drain in Nickel, i can still read in Koreader for a hour or two.